Priit Jaanson

A reference material with close to Lambertian reflectance and fluorescence emission profiles

Fluorescent brightening agents are widely used in various industries to enhance the appearance of materials. The angular profiles of emission and reflectance of fluorescent surfaces have been shown to deviate from Lambertian behaviour, however, in industry and calibration facilities single geometry measurements are often used, which requires assumptions to be made on the angular distributions. In addition, the angular distribution of reflectance has been shown to deviate from that of fluorescence. In this work, it is shown that the angular distribution of reflectance is dependent on the excitation wavelength and the effect is explained by qualitative and quantitative models. These angular and spectral effects may cause measurement errors when single geometry bidirectional measurements are carried out. The angular distributions can be taken into account by using goniometrical measurements, which however, result in increased calibration time and cost. Alternatively, a reference material could be used where the angular dependencies are minimised. In this work, a novel material is presented which demonstrates more Lambertian emission and reflectance profiles than conventional polytetrafluoroethylene (PTFE) based materials and a smaller dependence of angular reflectance on the absorbance of the sample.

Gonioreflectometric properties of metal surfaces

Angularly resolved measurements of scattered light from surfaces can provide useful information in various fields of research and industry, such as computer graphics, satellite based Earth observation etc. In practice, empirical or physics-based models are needed to interpolate the measurement results, because a thorough characterization of the surfaces under all relevant conditions may not be feasible. In this work, plain and anodized metal samples were prepared and measured optically for bidirectional reflectance distribution function (BRDF) and mechanically for surface roughness. Two models for BRDF (Torrance?Sparrow model and a polarimetric BRDF model) were fitted to the measured values. A better fit was obtained for plain metal surfaces than for anodized surfaces.

Characterization of thin-film thickness

Analysis methods and instrumentation for obtaining optical parameters and thickness profiles of thin-film samples from spectrophotometric and ellipsometric measurements are presented. Measured samples include thermally grown and evaporated SiO2 on a silicon substrate and a polymer photoresist layer on silicon. Experimental results at multiple sample positions give the thickness uniformity and optical constants of thin films. The thickness results obtained with spectrophotometry and ellipsometry agree within 1 nm for the 300 nm thick layer of SiO2 on silicon. For the 1600 nm thick resist sample the agreement of the measurement methods is within 8 nm. For the sample with a nominally 6000 nm thick layer of SiO2 on silicon, there is a deviation of ~100 nm between the spectrophotometry and ellipsometry results. As an application, the optical parameters of a SiO2 layer on an induced junction silicon photodiode are determined by spectrophotometry and are used to confirm earlier values and uncertainties of the SiO2 refractive index and layer thickness non-uniformity.

Toward SI Traceability of a Monte Carlo Radiative Transfer Model in the Visible Range

A 3-D Monte Carlo (MC) ray-tracing radiative transfer model is tested for its ability to simulate the bidirectional reflectance factors (BRFs) of a grooved artificial target given SI-traceable measurements of the optical and topographic properties of the target’s surface. The optical properties of a grooved target and an identical flat target were measured with the goniospectrophotometer at the National Metrology Institute of U.K. (NPL) and are traceable to the NPL scales of radiance factor. The topographic measurements were performed with the coordinate measuring machine at the National Metrology Institute of Finland (MIKES), and are traceable to the realization of the meter. The BRFs of the flat target were used to parameterize analytical scattering functions for rough surfaces. Similarly, the topographic measurement results were used to construct a structural model of the grooved target. Each element within this structural model then had its optical properties defined by the parameterized scattering function before the 3-D MC model simulated the BRFs of the grooved target under well-defined illumination and viewing conditions. The measured and modeled BRFs agreed for 72% of the measured geometries in the plane of incidence within the measurement and modeling uncertainties. The relative root-mean-squared (RMSE) error was 0.19. In the plane orthogonal to the plane of incidence, the measured and modeled BRFs agreed for 45% of the measured geometries, and the relative RMSE between measured and modeled values was 0.65.

Goniometrical measurements of fluorescence quantum efficiency

Fluorescent brightening agents are widely used in various industries to enhance the appearance of materials. In order to describe the colour of a fluorescent surface, accurate measurements of fluorescence emission are needed. In this work, measurements of goniometrically resolved bispectral luminescent radiance factors are presented with improved uncertainties. In addition, a method for goniometrically determining the spectral quantum efficiencies of the fluorescent process is presented and validated. The advantage of goniometric measurements is that they avoid the need to assume a specific angular emission profile of the fluorescent sample.

Improved diffusers for solar UV spectroradiometers

Diffuser heads whose angular response is proportional to the cosine of the zenith angle are needed for global irradiance measurements. Various material samples were measured for their transmission properties to find out the most promising material candidate for use in an improved solar UV diffuser. Novel bubbled quartz materials were found to be attractive alternatives to the traditional PTFE materials for this purpose. A 3D Monte Carlo particle tracing software was constructed and used to optimize the diffuser design. Integrated cosine error of 2.3% was measured with a raised flat diffuser on a prototype detector. The preliminary simulation results indicate that integrated cosine error can be lowered to 1.7 % or below simply by adjusting the dimensions of a flat diffuser. Integrated cosine error of 0.8 % or below can be reached with shaped diffusers.